Partie 1 : Etude de la fluidité (backbone
vs ISL1/2/2.1 vs ISL3/5 vs ISL4) sur la complétion > 90%
Importation des données
tree <- read.tree("/home/maarque/Donnees_projet/Data/SAGs_reduits/SAGs_reduits_reroot.nwk")
pa <- read_rtab("/home/maarque/Donnees_projet/Data/SAGs_reduits/matrix_SAGs_reduits.tab")
fit <- panstripe(pa, tree, family="quasipoisson")
Import de la matrice d’annotations
data_compartiments=read.csv("/home/maarque/Donnees_projet/Data/ALL_SAGS_genrefs/Annot_All_COGS.tsv", header=F, sep="\t", check.names = F)
colnames(data_compartiments)=c("COG","multi/single","gentype","shared","compartiment")
raw_data=read.csv("/home/maarque/Donnees_projet/Data/SAGs_reduits/Data_count/Count_BASH/family.txt", header=T, sep="\t", check.names = F)
data_count=raw_data[,c("name","Gains","Losses")]
colnames(data_count)[1]="COG"
data_merge_spec=merge(data_count,data_compartiments, by.x="COG", all = FALSE)
data_merge_spec=cbind(data_merge_spec,All_events=data_merge_spec$Gains+data_merge_spec$Losses)
Fluidité des compartiments core + flex
# Calcul de la fluidité moyenne par compartiment génétique
fluidite_par_compartiment <- tapply(data_merge_spec$All_events, data_merge_spec$compartiment, mean)
nombre_COGS_par_compartiment <- tapply(data_merge_spec$COG, data_merge_spec$compartiment, length)
# Créer un nouveau data frame avec les résultats
COGS_fluidite_par_compartiment <- data.frame(compartiment = names(fluidite_par_compartiment),
fluidite = fluidite_par_compartiment,
nombre_COGS = nombre_COGS_par_compartiment)
Différenciation core/flex
core=tapply(data_merge_spec$All_events[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
mean)
# Calculer la moyenne par compartiment pour le type "flex"
flex=tapply(data_merge_spec$All_events[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
# Compter le nombre de COGS par compartiment pour le type "core"
nombre_COGS_core=tapply(data_merge_spec$COG[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
length)
# Compter le nombre de COGS par compartiment pour le type "flex"
nombre_COGS_flex=tapply(data_merge_spec$COG[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
COGS_core <- data.frame(compartiment = names(core), fluidite = core, nombre_COGS = nombre_COGS_core)
COGS_flex <- data.frame(compartiment = names(flex), fluidite = flex, nombre_COGS = nombre_COGS_flex)
Résultats
#Rapport de fluidité par compartiments
print(fluidite_par_compartiment)
## Ambiguous Backbone comp ISL1 ISL2 ISL3 ISL4 ISL5
## 0.5304659 0.9868660 3.0526316 1.3500000 1.7627119 1.1183432 0.5914787 1.3920863
#Rapport de fluidité par compartiments core
print(core)
## Backbone ISL1 ISL2 ISL3 ISL4 ISL5
## 1.833716 2.000000 4.166667 3.545455 3.750000 3.187500
#Rapport de fluidité par compartiments flex
print(flex)
## Ambiguous Backbone ISL1 ISL2 ISL3 ISL4 ISL5
## 0.5304659 0.6665740 1.3461538 1.4210526 1.0207469 0.5500634 1.2226562
BAR Plots
melt_all=melt(fluidite_par_compartiment)
melt_all_clean=cbind(melt_all,COGS=t(t(COGS_fluidite_par_compartiment[,3])))
melt_all_clean=melt_all_clean[-c(1,3),]
melt_core=melt(core)
melt_core_clean=cbind(melt_core,COGS=t(t(COGS_core[,3])))
melt_flex=melt(flex)
melt_flex_clean=cbind(melt_flex,COGS=t(t(COGS_flex[,3])))
melt_flex_clean=melt_flex_clean[-c(1),]
ggplot(melt_all_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("Fluidité en fonction des compartiments pour le génome core + flex") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_core_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("Fluidité en fonction des compartiments pour le génome core") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_flex_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("Fluidité en fonction des compartiments pour le génome flex") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
VIOLIN Plots
# Supprimer les lignes contenant "comp" ou "Ambiguous" dans la colonne "compartiment"
data_merge_spec2=subset(data_merge_spec, !grepl("comp|Ambiguous", compartiment) & grepl("",compartiment))
# Pour gentype core + flex
ggplot(data_merge_spec2, aes(x = compartiment, y = All_events, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("Violin plot du nombre d'évènements par compartiments pour le génome CORE + FLEX") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
# Filtrer les données pour le gentype "core"
data_core=subset(data_merge_spec2, gentype == "core")
ggplot(data_core, aes(x = compartiment, y = All_events, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("Violin plot du nombre d'évènements par compartiments pour le génome CORE") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
## Warning: Groups with fewer than two data points have been dropped.
# Filtrer les données pour le gentype "flex"
data_flex=subset(data_merge_spec2, gentype == "flex")
ggplot(data_flex, aes(x = compartiment, y = All_events, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("Violin plot du nombre d'évènements par compartiments pour le génome FLEX") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
Fluidité des compartiments en gain vs
pertes
fluidite_par_compartiment_gain=tapply(data_merge_spec$Gains, data_merge_spec$compartiment, mean)
fluidite_par_compartiment_loss=tapply(data_merge_spec$Losses, data_merge_spec$compartiment, mean)
nombre_COGS_par_compartiment_gain <- tapply(data_merge_spec$Gains, data_merge_spec$compartiment, length)
nombre_COGS_par_compartiment_loss <- tapply(data_merge_spec$Losses, data_merge_spec$compartiment, length)
COGS_fluidite_par_compartiment_gain <- data.frame(compartiment = names(fluidite_par_compartiment_gain),
fluidite = fluidite_par_compartiment_gain,
nombre_COGS = nombre_COGS_par_compartiment_gain)
COGS_fluidite_par_compartiment_loss <- data.frame(compartiment = names(fluidite_par_compartiment_loss),
fluidite = fluidite_par_compartiment_loss,
nombre_COGS = nombre_COGS_par_compartiment_loss)
core_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
mean)
core_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
mean)
flex_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
flex_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
nombre_COGS_core_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
length)
nombre_COGS_core_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
length)
# Compter le nombre de COGS par compartiment pour le type "flex"
nombre_COGS_flex_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
nombre_COGS_flex_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
COGS_core_gain <- data.frame(compartiment = names(core_gain), fluidite = core_gain, nombre_COGS = nombre_COGS_core_gain)
COGS_core_loss <- data.frame(compartiment = names(core_loss), fluidite = core_loss, nombre_COGS = nombre_COGS_core_loss)
COGS_flex_gain <- data.frame(compartiment = names(flex_gain), fluidite = flex_gain, nombre_COGS = nombre_COGS_flex_gain)
COGS_flex_loss <- data.frame(compartiment = names(flex_loss), fluidite = flex_loss, nombre_COGS = nombre_COGS_flex_loss)
BAR Plots GAIN LOSS
melt_all_gain=melt(fluidite_par_compartiment_gain)
melt_all_gain_clean=cbind(melt_all_gain,COGS=t(t(COGS_fluidite_par_compartiment_gain[,3])))
melt_all_gain_clean=melt_all_gain_clean[-c(1,3),]
melt_all_loss=melt(fluidite_par_compartiment_loss)
melt_all_loss_clean=cbind(melt_all_loss,COGS=t(t(COGS_fluidite_par_compartiment_loss[,3])))
melt_all_loss_clean=melt_all_loss_clean[-c(1,3),]
melt_core_gain=melt(core_gain)
melt_core_gain_clean=cbind(melt_core_gain,COGS=t(t(COGS_core_gain[,3])))
melt_core_loss=melt(core_loss)
melt_core_loss_clean=cbind(melt_core_loss,COGS=t(t(COGS_core_loss[,3])))
melt_flex_gain=melt(flex_gain)
melt_flex_gain_clean=cbind(melt_flex_gain,COGS=t(t(COGS_flex_gain[,3])))
melt_flex_gain_clean=melt_flex_gain_clean[-c(1),]
melt_flex_loss=melt(flex_loss)
melt_flex_loss_clean=cbind(melt_flex_loss,COGS=t(t(COGS_flex_loss[,3])))
melt_flex_loss_clean=melt_flex_loss_clean[-c(1),]
ggplot(melt_all_gain_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("GAIN : Fluidité en fonction des compartiments pour le génome core + flex (18)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_all_loss_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("LOSS : Fluidité en fonction des compartiments pour le génome core + flex (18)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_core_gain_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("GAIN : Fluidité en fonction des compartiments pour le génome core (18)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_core_loss_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("LOSS : Fluidité en fonction des compartiments pour le génome core (18)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_flex_gain_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("GAIN : Fluidité en fonction des compartiments pour le génome flex (18)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_flex_loss_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("LOSS : Fluidité en fonction des compartiments pour le génome flex (18)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
VIOLIN Plots GAIN LOSS
# Supprimer les lignes contenant "comp" ou "Ambiguous" dans la colonne "compartiment"
data_merge_spec2=subset(data_merge_spec, !grepl("comp|Ambiguous", compartiment) & grepl("",compartiment))
# Pour gentype core + flex
ggplot(data_merge_spec2, aes(x = compartiment, y = Gains, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("GAINS : Violin plot du nombre d'évènements par compartiments pour le génome CORE + FLEX (18)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
ggplot(data_merge_spec2, aes(x = compartiment, y = Losses, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("LOSSES : Violin plot du nombre d'évènements par compartiments pour le génome CORE + FLEX (18)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
# Filtrer les données pour le gentype "core"
data_core=subset(data_merge_spec2, gentype == "core")
ggplot(data_core, aes(x = compartiment, y = Gains, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("GAINS : Violin plot du nombre d'évènements par compartiments pour le génome CORE (18)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
## Warning: Groups with fewer than two data points have been dropped.
ggplot(data_core, aes(x = compartiment, y = Losses, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("LOSSES : Violin plot du nombre d'évènements par compartiments pour le génome CORE (18)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
## Warning: Groups with fewer than two data points have been dropped.
# Filtrer les données pour le gentype "flex"
data_flex=subset(data_merge_spec2, gentype == "flex")
ggplot(data_flex, aes(x = compartiment, y = Gains, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("GAINS : Violin plot du nombre d'évènements par compartiments pour le génome FLEX (18)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
ggplot(data_flex, aes(x = compartiment, y = Losses, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("LOSSES : Violin plot du nombre d'évènements par compartiments pour le génome FLEX (18)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
Partie 2 : Phylogénie et fluidité des 7
génomes de références HLI et HLII
Importation des données
HLI=c("MED4","MIT9515")
HLII=c("P9312","MIT9215","AS9601","MIT9202","MIT9301")
HL=c(HLI,HLII)
tree=read.tree("/home/maarque/Donnees_projet/Data/ALL_SAGS_genrefs/REROOT_RAxML_bestTree.ALL_concatenated_cogs.nwk")
pa=read_rtab("/home/maarque/Donnees_projet/Data/ALL_SAGS_genrefs/Matrice_fusionnee.tsv")
tree = KeepTip(tree, HL, preorder = TRUE)
pa=pa[HL,]
Utilisation de panstripe
fit <- panstripe(pa, tree, family="quasipoisson")
Plot du gain/perte
gt <- dplyr::full_join(ggtree::fortify(fit$tree), data.frame(node = fit$tree$edge[,2], trait = fit$data$acc), by = "node")
data_x=gt$x
data_y=gt$y
data_labels=round(gt$trait)
#Arbre gain/perte qui affiche les gains/pertes de gènes
plot_gain_loss(fit,tip_label = FALSE)+geom_tiplab(align=TRUE,size=1.2,color="black")+ geom_text(aes(data_x,data_y),label=data_labels, check_overlap = FALSE, color="red", size=1.5,hjust=-0.3,vjust=-0.1)
## Warning: Removed 1 rows containing missing values (`geom_text()`).
Annotations des COGS + importation données
COUNT
data_compartiments=read.csv("/home/maarque/Donnees_projet/Data/ALL_SAGS_genrefs/Annot_All_COGS.tsv", header=F, sep="\t", check.names = F)
colnames(data_compartiments)=c("COG","multi/single","gentype","shared","compartiment")
raw_data=read.csv("/home/maarque/Donnees_projet/Data/7_Genrefs/COUNT_bash/family.txt", header=T, sep="\t", check.names = F)
data_count=raw_data[,c("name","Gains","Losses")]
colnames(data_count)[1]="COG"
data_merge_spec=merge(data_count,data_compartiments, by.x="COG", all = FALSE)
data_merge_spec=cbind(data_merge_spec,All_events=data_merge_spec$Gains+data_merge_spec$Losses)
Fluidité des compartiments core + flex
# Calcul de la fluidité moyenne par compartiment génétique
fluidite_par_compartiment <- tapply(data_merge_spec$All_events, data_merge_spec$compartiment, mean)
nombre_COGS_par_compartiment <- tapply(data_merge_spec$COG, data_merge_spec$compartiment, length)
# Créer un nouveau data frame avec les résultats
COGS_fluidite_par_compartiment <- data.frame(compartiment = names(fluidite_par_compartiment),
fluidite = fluidite_par_compartiment,
nombre_COGS = nombre_COGS_par_compartiment)
Différenciation core/flex
core=tapply(data_merge_spec$All_events[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
mean)
# Calculer la moyenne par compartiment pour le type "flex"
flex=tapply(data_merge_spec$All_events[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
# Compter le nombre de COGS par compartiment pour le type "core"
nombre_COGS_core=tapply(data_merge_spec$COG[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
length)
# Compter le nombre de COGS par compartiment pour le type "flex"
nombre_COGS_flex=tapply(data_merge_spec$COG[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
COGS_core <- data.frame(compartiment = names(core), fluidite = core, nombre_COGS = nombre_COGS_core)
COGS_flex <- data.frame(compartiment = names(flex), fluidite = flex, nombre_COGS = nombre_COGS_flex)
Résultats
#Rapport de fluidité par compartiments
print(fluidite_par_compartiment)
## Ambiguous Backbone comp ISL1 ISL2 ISL3 ISL4
## 0.15770609 0.08365326 0.80701754 0.31250000 0.33333333 0.29980276 0.19172932
## ISL5
## 0.24100719
#Rapport de fluidité par compartiments core
print(core)
## Backbone ISL1 ISL2 ISL3 ISL4 ISL5
## 0.1348659 1.0000000 0.3333333 0.7272727 0.2500000 0.2500000
#Rapport de fluidité par compartiments flex
print(flex)
## Ambiguous Backbone ISL1 ISL2 ISL3 ISL4 ISL5
## 0.15770609 0.06390664 0.30769231 0.31578947 0.26763485 0.18504436 0.19921875
BAR Plots
melt_all=melt(fluidite_par_compartiment)
melt_all_clean=cbind(melt_all,COGS=t(t(COGS_fluidite_par_compartiment[,3])))
melt_all_clean=melt_all_clean[-c(1,3),]
melt_core=melt(core)
melt_core_clean=cbind(melt_core,COGS=t(t(COGS_core[,3])))
melt_flex=melt(flex)
melt_flex_clean=cbind(melt_flex,COGS=t(t(COGS_flex[,3])))
melt_flex_clean=melt_flex_clean[-c(1),]
ggplot(melt_all_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("Fluidité en fonction des compartiments pour le génome core + flex") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_core_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("Fluidité en fonction des compartiments pour le génome core") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_flex_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("Fluidité en fonction des compartiments pour le génome flex") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
VIOLIN Plots
# Supprimer les lignes contenant "comp" ou "Ambiguous" dans la colonne "compartiment"
data_merge_spec2=subset(data_merge_spec, !grepl("comp|Ambiguous", compartiment) & grepl("",compartiment))
# Pour gentype core + flex
ggplot(data_merge_spec2, aes(x = compartiment, y = All_events, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("Violin plot du nombre d'évènements par compartiments pour le génome CORE + FLEX") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
# Filtrer les données pour le gentype "core"
data_core=subset(data_merge_spec2, gentype == "core")
ggplot(data_core, aes(x = compartiment, y = All_events, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("Violin plot du nombre d'évènements par compartiments pour le génome CORE") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
## Warning: Groups with fewer than two data points have been dropped.
# Filtrer les données pour le gentype "flex"
data_flex=subset(data_merge_spec2, gentype == "flex")
ggplot(data_flex, aes(x = compartiment, y = All_events, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("Violin plot du nombre d'évènements par compartiments pour le génome FLEX") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
Fluidité des compartiments en gain vs
pertes
fluidite_par_compartiment_gain=tapply(data_merge_spec$Gains, data_merge_spec$compartiment, mean)
fluidite_par_compartiment_loss=tapply(data_merge_spec$Losses, data_merge_spec$compartiment, mean)
nombre_COGS_par_compartiment_gain <- tapply(data_merge_spec$Gains, data_merge_spec$compartiment, length)
nombre_COGS_par_compartiment_loss <- tapply(data_merge_spec$Losses, data_merge_spec$compartiment, length)
COGS_fluidite_par_compartiment_gain <- data.frame(compartiment = names(fluidite_par_compartiment_gain),
fluidite = fluidite_par_compartiment_gain,
nombre_COGS = nombre_COGS_par_compartiment_gain)
COGS_fluidite_par_compartiment_loss <- data.frame(compartiment = names(fluidite_par_compartiment_loss),
fluidite = fluidite_par_compartiment_loss,
nombre_COGS = nombre_COGS_par_compartiment_loss)
core_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
mean)
core_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
mean)
flex_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
flex_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
nombre_COGS_core_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
length)
nombre_COGS_core_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
length)
# Compter le nombre de COGS par compartiment pour le type "flex"
nombre_COGS_flex_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
nombre_COGS_flex_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
COGS_core_gain <- data.frame(compartiment = names(core_gain), fluidite = core_gain, nombre_COGS = nombre_COGS_core_gain)
COGS_core_loss <- data.frame(compartiment = names(core_loss), fluidite = core_loss, nombre_COGS = nombre_COGS_core_loss)
COGS_flex_gain <- data.frame(compartiment = names(flex_gain), fluidite = flex_gain, nombre_COGS = nombre_COGS_flex_gain)
COGS_flex_loss <- data.frame(compartiment = names(flex_loss), fluidite = flex_loss, nombre_COGS = nombre_COGS_flex_loss)
BAR Plots GAIN LOSS
melt_all_gain=melt(fluidite_par_compartiment_gain)
melt_all_gain_clean=cbind(melt_all_gain,COGS=t(t(COGS_fluidite_par_compartiment_gain[,3])))
melt_all_gain_clean=melt_all_gain_clean[-c(1,3),]
melt_all_loss=melt(fluidite_par_compartiment_loss)
melt_all_loss_clean=cbind(melt_all_loss,COGS=t(t(COGS_fluidite_par_compartiment_loss[,3])))
melt_all_loss_clean=melt_all_loss_clean[-c(1,3),]
melt_core_gain=melt(core_gain)
melt_core_gain_clean=cbind(melt_core_gain,COGS=t(t(COGS_core_gain[,3])))
melt_core_loss=melt(core_loss)
melt_core_loss_clean=cbind(melt_core_loss,COGS=t(t(COGS_core_loss[,3])))
melt_flex_gain=melt(flex_gain)
melt_flex_gain_clean=cbind(melt_flex_gain,COGS=t(t(COGS_flex_gain[,3])))
melt_flex_gain_clean=melt_flex_gain_clean[-c(1),]
melt_flex_loss=melt(flex_loss)
melt_flex_loss_clean=cbind(melt_flex_loss,COGS=t(t(COGS_flex_loss[,3])))
melt_flex_loss_clean=melt_flex_loss_clean[-c(1),]
ggplot(melt_all_gain_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("GAIN : Fluidité en fonction des compartiments pour le génome core + flex (7)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_all_loss_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("LOSS : Fluidité en fonction des compartiments pour le génome core + flex (7)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_core_gain_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("GAIN : Fluidité en fonction des compartiments pour le génome core (7)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_core_loss_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("LOSS : Fluidité en fonction des compartiments pour le génome core (7)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_flex_gain_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("GAIN : Fluidité en fonction des compartiments pour le génome flex (7)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_flex_loss_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("LOSS : Fluidité en fonction des compartiments pour le génome flex (7)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
VIOLIN Plots GAIN LOSS
# Supprimer les lignes contenant "comp" ou "Ambiguous" dans la colonne "compartiment"
data_merge_spec2=subset(data_merge_spec, !grepl("comp|Ambiguous", compartiment) & grepl("",compartiment))
# Pour gentype core + flex
ggplot(data_merge_spec2, aes(x = compartiment, y = Gains, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("GAINS : Violin plot du nombre d'évènements par compartiments pour le génome CORE + FLEX (7)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
ggplot(data_merge_spec2, aes(x = compartiment, y = Losses, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("LOSSES : Violin plot du nombre d'évènements par compartiments pour le génome CORE + FLEX (7)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
# Filtrer les données pour le gentype "core"
data_core=subset(data_merge_spec2, gentype == "core")
ggplot(data_core, aes(x = compartiment, y = Gains, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("GAINS : Violin plot du nombre d'évènements par compartiments pour le génome CORE (7)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
## Warning: Groups with fewer than two data points have been dropped.
ggplot(data_core, aes(x = compartiment, y = Losses, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("LOSSES : Violin plot du nombre d'évènements par compartiments pour le génome CORE (7)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
## Warning: Groups with fewer than two data points have been dropped.
# Filtrer les données pour le gentype "flex"
data_flex=subset(data_merge_spec2, gentype == "flex")
ggplot(data_flex, aes(x = compartiment, y = Gains, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("GAINS : Violin plot du nombre d'évènements par compartiments pour le génome FLEX (7)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
ggplot(data_flex, aes(x = compartiment, y = Losses, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("LOSSES : Violin plot du nombre d'évènements par compartiments pour le génome FLEX (7)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
Partie 3 : Phylogénie et fluidité des 7
génomes de références HLI et HLII + 18 SAGS
Importation des données
HLI=c("MED4","MIT9515")
HLII=c("P9312","MIT9215","AS9601","MIT9202","MIT9301")
SAGs=c("496N4_C1","521B10_C1","498P15_C1","529C4_C1","495K23_C1","498B22_C2","498C16_C2","529D18_C3","496A2_C3","529J15_C3","518A17_C3","495L20_C3","528N17_C4","498I20_C5","498A3_C8","527L22_C8","520K10_C9","528J8_C9")
All_data=c(HLI,HLII,SAGs)
tree=read.tree("/home/maarque/Donnees_projet/Data/ALL_SAGS_genrefs/REROOT_RAxML_bestTree.ALL_concatenated_cogs.nwk")
pa=read_rtab("/home/maarque/Donnees_projet/Data/ALL_SAGS_genrefs/Matrice_fusionnee.tsv")
tree = KeepTip(tree, All_data, preorder = TRUE)
pa=pa[All_data,]
Utilisation de panstripe
fit <- panstripe(pa, tree, family="quasipoisson")
Plot du gain/perte
gt <- dplyr::full_join(ggtree::fortify(fit$tree), data.frame(node = fit$tree$edge[,2], trait = fit$data$acc), by = "node")
data_x=gt$x
data_y=gt$y
data_labels=round(gt$trait)
#Arbre gain/perte qui affiche les gains/pertes de gènes
plot_gain_loss(fit,tip_label = FALSE)+geom_tiplab(align=TRUE,size=1.2,color="black")+ geom_text(aes(data_x,data_y),label=data_labels, check_overlap = FALSE, color="red", size=1.5,hjust=-0.3,vjust=-0.1)
## Warning: Removed 1 rows containing missing values (`geom_text()`).
Annotations des COGS + importation données
COUNT
data_compartiments=read.csv("/home/maarque/Donnees_projet/Data/ALL_SAGS_genrefs/Annot_All_COGS.tsv", header=F, sep="\t", check.names = F)
colnames(data_compartiments)=c("COG","multi/single","gentype","shared","compartiment")
raw_data=read.csv("/home/maarque/Donnees_projet/Data/7_Genrefs_18_SAGS/COUNT_bash/family.txt", header=T, sep="\t", check.names = F)
data_count=raw_data[,c("name","Gains","Losses")]
colnames(data_count)[1]="COG"
data_merge_spec=merge(data_count,data_compartiments, by.x="COG", all = FALSE)
data_merge_spec=cbind(data_merge_spec,All_events=data_merge_spec$Gains+data_merge_spec$Losses)
Fluidité des compartiments core + flex
# Calcul de la fluidité moyenne par compartiment génétique
fluidite_par_compartiment <- tapply(data_merge_spec$All_events, data_merge_spec$compartiment, mean)
nombre_COGS_par_compartiment <- tapply(data_merge_spec$COG, data_merge_spec$compartiment, length)
# Créer un nouveau data frame avec les résultats
COGS_fluidite_par_compartiment <- data.frame(compartiment = names(fluidite_par_compartiment),
fluidite = fluidite_par_compartiment,
nombre_COGS = nombre_COGS_par_compartiment)
Différenciation core/flex
core=tapply(data_merge_spec$All_events[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
mean)
# Calculer la moyenne par compartiment pour le type "flex"
flex=tapply(data_merge_spec$All_events[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
# Compter le nombre de COGS par compartiment pour le type "core"
nombre_COGS_core=tapply(data_merge_spec$COG[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
length)
# Compter le nombre de COGS par compartiment pour le type "flex"
nombre_COGS_flex=tapply(data_merge_spec$COG[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
COGS_core <- data.frame(compartiment = names(core), fluidite = core, nombre_COGS = nombre_COGS_core)
COGS_flex <- data.frame(compartiment = names(flex), fluidite = flex, nombre_COGS = nombre_COGS_flex)
Résultats
#Rapport de fluidité par compartiments
print(fluidite_par_compartiment)
## Ambiguous Backbone comp ISL1 ISL2 ISL3 ISL4 ISL5
## 0.1612903 0.6651849 1.3859649 0.4750000 0.7005650 0.4299803 0.2205514 0.4712230
#Rapport de fluidité par compartiments core
print(core)
## Backbone ISL1 ISL2 ISL3 ISL4 ISL5
## 1.989272 3.000000 2.000000 1.909091 2.250000 2.000000
#Rapport de fluidité par compartiments flex
print(flex)
## Ambiguous Backbone ISL1 ISL2 ISL3 ISL4 ISL5
## 0.1612903 0.1675465 0.4230769 0.4736842 0.3464730 0.1951838 0.3085938
BAR Plots
melt_all=melt(fluidite_par_compartiment)
melt_all_clean=cbind(melt_all,COGS=t(t(COGS_fluidite_par_compartiment[,3])))
melt_all_clean=melt_all_clean[-c(1,3),]
melt_core=melt(core)
melt_core_clean=cbind(melt_core,COGS=t(t(COGS_core[,3])))
melt_flex=melt(flex)
melt_flex_clean=cbind(melt_flex,COGS=t(t(COGS_flex[,3])))
melt_flex_clean=melt_flex_clean[-c(1),]
ggplot(melt_all_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("Fluidité en fonction des compartiments pour le génome core + flex") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_core_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("Fluidité en fonction des compartiments pour le génome core") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_flex_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("Fluidité en fonction des compartiments pour le génome flex") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
VIOLIN Plots
# Supprimer les lignes contenant "comp" ou "Ambiguous" dans la colonne "compartiment"
data_merge_spec2=subset(data_merge_spec, !grepl("comp|Ambiguous", compartiment) & grepl("",compartiment))
# Pour gentype core + flex
ggplot(data_merge_spec2, aes(x = compartiment, y = All_events, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("Violin plot du nombre d'évènements par compartiments pour le génome CORE + FLEX") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
# Filtrer les données pour le gentype "core"
data_core=subset(data_merge_spec2, gentype == "core")
ggplot(data_core, aes(x = compartiment, y = All_events, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("Violin plot du nombre d'évènements par compartiments pour le génome CORE") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
## Warning: Groups with fewer than two data points have been dropped.
# Filtrer les données pour le gentype "flex"
data_flex=subset(data_merge_spec2, gentype == "flex")
ggplot(data_flex, aes(x = compartiment, y = All_events, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("Violin plot du nombre d'évènements par compartiments pour le génome FLEX") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
Fluidité des compartiments en gain vs
pertes
fluidite_par_compartiment_gain=tapply(data_merge_spec$Gains, data_merge_spec$compartiment, mean)
fluidite_par_compartiment_loss=tapply(data_merge_spec$Losses, data_merge_spec$compartiment, mean)
nombre_COGS_par_compartiment_gain <- tapply(data_merge_spec$Gains, data_merge_spec$compartiment, length)
nombre_COGS_par_compartiment_loss <- tapply(data_merge_spec$Losses, data_merge_spec$compartiment, length)
COGS_fluidite_par_compartiment_gain <- data.frame(compartiment = names(fluidite_par_compartiment_gain),
fluidite = fluidite_par_compartiment_gain,
nombre_COGS = nombre_COGS_par_compartiment_gain)
COGS_fluidite_par_compartiment_loss <- data.frame(compartiment = names(fluidite_par_compartiment_loss),
fluidite = fluidite_par_compartiment_loss,
nombre_COGS = nombre_COGS_par_compartiment_loss)
core_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
mean)
core_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
mean)
flex_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
flex_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
nombre_COGS_core_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
length)
nombre_COGS_core_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "core"],
data_merge_spec$compartiment[data_merge_spec$gentype == "core"],
length)
# Compter le nombre de COGS par compartiment pour le type "flex"
nombre_COGS_flex_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
nombre_COGS_flex_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
COGS_core_gain <- data.frame(compartiment = names(core_gain), fluidite = core_gain, nombre_COGS = nombre_COGS_core_gain)
COGS_core_loss <- data.frame(compartiment = names(core_loss), fluidite = core_loss, nombre_COGS = nombre_COGS_core_loss)
COGS_flex_gain <- data.frame(compartiment = names(flex_gain), fluidite = flex_gain, nombre_COGS = nombre_COGS_flex_gain)
COGS_flex_loss <- data.frame(compartiment = names(flex_loss), fluidite = flex_loss, nombre_COGS = nombre_COGS_flex_loss)
BAR Plots GAIN LOSS
melt_all_gain=melt(fluidite_par_compartiment_gain)
melt_all_gain_clean=cbind(melt_all_gain,COGS=t(t(COGS_fluidite_par_compartiment_gain[,3])))
melt_all_gain_clean=melt_all_gain_clean[-c(1,3),]
melt_all_loss=melt(fluidite_par_compartiment_loss)
melt_all_loss_clean=cbind(melt_all_loss,COGS=t(t(COGS_fluidite_par_compartiment_loss[,3])))
melt_all_loss_clean=melt_all_loss_clean[-c(1,3),]
melt_core_gain=melt(core_gain)
melt_core_gain_clean=cbind(melt_core_gain,COGS=t(t(COGS_core_gain[,3])))
melt_core_loss=melt(core_loss)
melt_core_loss_clean=cbind(melt_core_loss,COGS=t(t(COGS_core_loss[,3])))
melt_flex_gain=melt(flex_gain)
melt_flex_gain_clean=cbind(melt_flex_gain,COGS=t(t(COGS_flex_gain[,3])))
melt_flex_gain_clean=melt_flex_gain_clean[-c(1),]
melt_flex_loss=melt(flex_loss)
melt_flex_loss_clean=cbind(melt_flex_loss,COGS=t(t(COGS_flex_loss[,3])))
melt_flex_loss_clean=melt_flex_loss_clean[-c(1),]
ggplot(melt_all_gain_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("GAIN : Fluidité en fonction des compartiments pour le génome core + flex (25)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_all_loss_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("LOSS : Fluidité en fonction des compartiments pour le génome core + flex (25)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_core_gain_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("GAIN : Fluidité en fonction des compartiments pour le génome core (25)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_core_loss_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("LOSS : Fluidité en fonction des compartiments pour le génome core (25)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_flex_gain_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("GAIN : Fluidité en fonction des compartiments pour le génome flex (25)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_flex_loss_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("LOSS : Fluidité en fonction des compartiments pour le génome flex (25)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
VIOLIN Plots GAIN LOSS
# Supprimer les lignes contenant "comp" ou "Ambiguous" dans la colonne "compartiment"
data_merge_spec2=subset(data_merge_spec, !grepl("comp|Ambiguous", compartiment) & grepl("",compartiment))
# Pour gentype core + flex
ggplot(data_merge_spec2, aes(x = compartiment, y = Gains, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("GAINS : Violin plot du nombre d'évènements par compartiments pour le génome CORE + FLEX (25)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
ggplot(data_merge_spec2, aes(x = compartiment, y = Losses, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("LOSSES : Violin plot du nombre d'évènements par compartiments pour le génome CORE + FLEX (25)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
# Filtrer les données pour le gentype "core"
data_core=subset(data_merge_spec2, gentype == "core")
ggplot(data_core, aes(x = compartiment, y = Gains, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("GAINS : Violin plot du nombre d'évènements par compartiments pour le génome CORE (25)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
## Warning: Groups with fewer than two data points have been dropped.
ggplot(data_core, aes(x = compartiment, y = Losses, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("LOSSES : Violin plot du nombre d'évènements par compartiments pour le génome CORE (25)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
## Warning: Groups with fewer than two data points have been dropped.
# Filtrer les données pour le gentype "flex"
data_flex=subset(data_merge_spec2, gentype == "flex")
ggplot(data_flex, aes(x = compartiment, y = Gains, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("GAINS : Violin plot du nombre d'évènements par compartiments pour le génome FLEX (25)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
ggplot(data_flex, aes(x = compartiment, y = Losses, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("LOSSES : Violin plot du nombre d'évènements par compartiments pour le génome FLEX (25)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
Partie 4 : Etude de la fluidité (backbone
vs ISL1/2/2.1 vs ISL3/5 vs ISL4) sur la complétion > 90% avec
uniquement les COGS flex des genrefs
Importation des données
tree <- read.tree("/home/maarque/Donnees_projet/Data/SAGs_reduits/SAGs_reduits_reroot.nwk")
pa <- read_rtab("/home/maarque/Donnees_projet/Data/SAGs_reduits/matrix_SAGs_reduits.tab")
fit <- panstripe(pa, tree, family="quasipoisson")
Import de la matrice d’annotations
data_compartiments=read.csv("/home/maarque/Donnees_projet/Data/Data_GARDON/Annotation_COGs/All_COGS_max.tsv", header=F, sep="\t", check.names = F)
colnames(data_compartiments)=c("COG","multi/single","gentype","shared","compartiment")
raw_data=read.csv("/home/maarque/Donnees_projet/Data/SAGs_reduits/Data_count/Count_BASH/family.txt", header=T, sep="\t", check.names = F)
data_count=raw_data[,c("name","Gains","Losses")]
colnames(data_count)[1]="COG"
data_merge_spec=merge(data_count,data_compartiments, by.x="COG", all = FALSE)
data_merge_spec=cbind(data_merge_spec,All_events=data_merge_spec$Gains+data_merge_spec$Losses)
data_merge_spec=subset(data_merge_spec, grepl("flex", gentype) & (grepl("shared w/ MIT9312", shared) | grepl("shared w/ HLII", shared)))
Récupération des données flex
# Calculer la moyenne par compartiment pour le type "flex"
flex=tapply(data_merge_spec$All_events[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
# Compter le nombre de COGS par compartiment pour le type "flex"
nombre_COGS_flex=tapply(data_merge_spec$COG[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
COGS_flex <- data.frame(compartiment = names(flex), fluidite = flex, nombre_COGS = nombre_COGS_flex)
Résultats
print(flex)
## Ambiguous Backbone ISL1 ISL2 ISL3 ISL4 ISL5
## 1.117647 2.012048 1.806452 3.346939 2.323944 1.361290 3.184615
BAR Plots
melt_flex=melt(flex)
melt_flex_clean=cbind(melt_flex,COGS=t(t(COGS_flex[,3])))
melt_flex_clean=melt_flex_clean[-c(1),]
ggplot(melt_flex_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("Fluidité en fonction des compartiments pour le génome flex") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
VIOLIN Plots
# Filtrer les données pour le gentype "flex"
data_flex=subset(data_merge_spec2, gentype == "flex")
ggplot(data_flex, aes(x = compartiment, y = All_events, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("Violin plot du nombre d'évènements par compartiments pour le génome FLEX") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
Fluidité des compartiments en gain vs
pertes
flex_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
flex_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
mean)
# Compter le nombre de COGS par compartiment pour le type "flex"
nombre_COGS_flex_gain=tapply(data_merge_spec$Gains[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
nombre_COGS_flex_loss=tapply(data_merge_spec$Losses[data_merge_spec$gentype == "flex"],
data_merge_spec$compartiment[data_merge_spec$gentype == "flex"],
length)
COGS_flex_gain <- data.frame(compartiment = names(flex_gain), fluidite = flex_gain, nombre_COGS = nombre_COGS_flex_gain)
COGS_flex_loss <- data.frame(compartiment = names(flex_loss), fluidite = flex_loss, nombre_COGS = nombre_COGS_flex_loss)
BAR Plots GAIN LOSS
melt_flex_gain=melt(flex_gain)
melt_flex_gain_clean=cbind(melt_flex_gain,COGS=t(t(COGS_flex_gain[,3])))
melt_flex_gain_clean=melt_flex_gain_clean[-c(1),]
melt_flex_loss=melt(flex_loss)
melt_flex_loss_clean=cbind(melt_flex_loss,COGS=t(t(COGS_flex_loss[,3])))
melt_flex_loss_clean=melt_flex_loss_clean[-c(1),]
ggplot(melt_flex_gain_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("GAIN : Fluidité en fonction des compartiments pour le génome flex (18)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
ggplot(melt_flex_loss_clean, aes(x = Var1, y = value, fill= Var1)) +
geom_bar(stat = "identity") +
xlab("Compartiment") +
ylab("Fluidité") +
ggtitle("LOSS : Fluidité en fonction des compartiments pour le génome flex (18)") +
theme_minimal()+
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
geom_text(aes(label = COGS), y=0.010, color="white" , size = 3) +
geom_text(aes(label = round(value, 2)), vjust = -0.5, size=3) + labs(fill="Légende :")
VIOLIN Plots GAIN LOSS
data_merge_spec2=subset(data_merge_spec, !grepl("comp|Ambiguous", compartiment) & grepl("",compartiment))
# Filtrer les données pour le gentype "flex"
data_flex=subset(data_merge_spec2, gentype == "flex")
ggplot(data_flex, aes(x = compartiment, y = Gains, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("GAINS : Violin plot du nombre d'évènements par compartiments pour le génome FLEX (18)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
ggplot(data_flex, aes(x = compartiment, y = Losses, fill = compartiment)) +
geom_violin() +
xlab("Compartiments") +
ylab("Nombre d'évènements") +
ggtitle("LOSSES : Violin plot du nombre d'évènements par compartiments pour le génome FLEX (18)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))+
ylim(0,8)
